The primary focus of my laboratory is to determine molecular mechanisms involved in prostate and mammary tumorigenesis using transgenic mouse approaches, and to use these animal models as systems in which to test novel therapies. A primary question is understanding what molecular events are involved in tumor progression. To this end, we have concentrated much effort on correlating the histogenesis of mammary and prostate lesions to molecular alterations that occur during the multistep process of carcinogenesis using the C3(1)/Tag transgenic model developed in our lab. Male C3(1)/Tag transgenic mice develop prostatic intraepithelial neoplasia (PIN) lesions very similar to those observed in humans, which often progress to invasive adenocarcinomas over several months. All female mice carrying the C3(1)/Tag transgene develop mammary adenocarcinomas over several months in a very predictable manner demonstrating transition lesions similar to DCIS found during human breast cancer development. We are using this unique transgenic model to compare mechanisms of tumorigenesis in two different hormone-dependent tissues within the same genetic background. We have expanded these studies to include models in which the overexpression of oncogenes utilizing different oncogenic pathways have been compared through gene expression profiling. Insights gained from these studies have demonstrated that mammary cancer initiated by different oncogenic mechanisms dysregulate gene expression in an overwhelmingly similar fashion. However, gene expression signatures related to specific molecular pathways can be identified which characterize each oncogene. We are now comparing tumor stage-specific differences in gene expression profiles using laser capture microscopy and microarray techniques. Studies are currently underway to better compare microarray data from mouse tumor models to that of human breast cancer in an effort to understand important biological similarities and differences in oncogenesis between the two different species. This approach is also a new and important aspect of model validation. Ongoing work is attempting to correlate changes at the genomic level in mouse tumor models with changes in gene expression, especially as it relates to human breast cancer. This includes the use of comparative genomic hybridization, spectral karyotyping, and BAC arrays. Important changes in the expression of genes that regulate the cell cycle have been identified, in particular, the loss of p21. Recent gene therapy approaches in our lab have demonstrated that the restoration of p21 function can significantly reduce mammary tumor progression using this transgenic model. We have also demonstrated that (I)bax(/I) expression is critical to protective apoptosis primarily during pre-neoplasia. Double transgenic mice lacking (I)bax(I) have a significantly accelerated progression of mammary tumors. New efforts are underway to understand molecular changes associated with metastases through the use of laser capture microscopy and microarray technologies. Emphasis is being placed on the development of new (I)in vivo (I) models of metastases with relevance to human breast and prostate cancer. Our lab is using various genetically-modified mice to assess chemopreventive and chemotherapeutic compounds. We have demonstrated that several chemopreventive agents can reduce mammary tumor incidence, multiplicity and growth. Mammary tumor growth in the C3(1)/Tag model has been inhibited by two different forms of recombinant endostatin as well as endostatin produced by adenovirus. Promising results are currently being obtained using 2-methoxyestradiol and a tyrosine kinase inhibitor of VEGF receptor. Our ongoing work is focused on applying expression profiling to identify molecular mechanisms of action of these tumor-inhibitory compounds as well as potential new targets of therapy.